Skip to main content

Table 3 Characteristics of RCTs with a positive response and negative or no response to the request for data sharing

From: Feasibility of individual patient data meta-analyses in orthopaedic surgery

Characteristics Data sharing requesta (n = 263) P value
n (%)
Positive response Negative or no response
37 (14%) 226 (86%)
RCT characteristics    
Study design    0.07
 Single-centre (n = 176) 19 (11%) 157 (89%)
 Multicentre (n = 57) 11 (19%) 46 (81%)
 Not reported (n = 30) 7 (23%) 23 (77%)
Location    0.11
 Europe (n = 105) 20 (19%) 85 (81%)
 North America (n = 71) 9 (13%) 62 (87%)
 Asia (n = 78) 6 (8%) 72 (92%)
 Australia and New Zealand (n = 9) 2 (22%) 7 (78%)
Funding    0.19
 Public (n = 91) 8 (9%) 83 (91%)  
 Privateb (n = 45) 7 (16%) 38 (84%)
 No response (n = 127) 22 (17%) 105 (83%)
Year of publication    0.57
 2000–2004 (n = 44) 4 (9%) 40 (91%)
 2005–2009 (n = 109) 16 (15%) 93 (85%)
 2010–2013 (n = 110) 17 (15%) 93 (85%)
Journal of publication characteristics    
Journal of publication    0.68
 Specialized (n = 251) 35 (14%) 216 (86%)
 Generalist (n = 12) 2 (17%) 10 (83%)
Journal data sharing policies    0.27
 Yes (support or mandatory) (n = 30) 2 (7%) 28 (93%)
 No (n = 233) 35 (15%) 198 (85%)
Top 10 impact factor in each specialty    0.21
 Yes (n = 117) 20 (17%) 97 (83%)
 No (n = 146) 17 (12%) 129 (88%)
  1. aTen trials excluded because our request was being evaluated by the research team or sponsor at the time of statistical analysis; btotal number (n = ) used for calculating the percentage